A potential of prognostic 7-genes expression signatures for predicts survival in patient with colorectal cancer

The aim of this study is to identify expression-based genes features with genetic sequence data in colorectal cancer (CR) patients. We collected genes expression profiling of CR from TCGA and GEO databases. Univariate and multivariate Cox regression analysis were used to evaluate the prognostic function of 7-genes siguatures based on risk score. We constructed the risk score model based on the expression data of 7-genes signatures in CRC train data set, and then validated in the test data set and TCGA database through bioinformatics analysis. We identified 7-genes, which were associated with the prognosis of 2 GEO cohorts in 307 CRC patients ( p < 0. 01). Patients in train data set could be divided into high-risk and low-risk groups with significantly different Overall survival (OS) and Kaplan-Meier (KM) based on 7-genes signatures. The signatures of the 7-genes were successfully validated in the independent TCGA test cohort and showed excellent performance in risk stratification associated with existing gene-related traits. Functional enrichment analyses revealed potential functional roles of these different expression genes in tumorigenesis. Our research demonstrated that the novel 7-genes signatures could be used as a potential independent biomarker for predicting the survival of CR patient.